Jiangxz commited on
Commit
8e2dc9e
·
verified ·
1 Parent(s): 131ab57

Update app.py

Browse files

Taiwan_Tax_Knowledge-base

Files changed (1) hide show
  1. app.py +4 -3
app.py CHANGED
@@ -10,7 +10,7 @@ from langchain_groq import ChatGroq
10
  from langchain.text_splitter import RecursiveCharacterTextSplitter
11
  from langchain_huggingface import HuggingFaceEmbeddings
12
  from langchain_community.vectorstores import Chroma
13
- from langchain.embeddings import HuggingFaceEmbeddings
14
  from langchain.chains import RetrievalQA
15
  from langchain_community.document_loaders import WebBaseLoader, TextLoader
16
  from langchain.prompts import PromptTemplate
@@ -74,18 +74,19 @@ text_splitter = RecursiveCharacterTextSplitter(
74
  )
75
 
76
  split_docs = text_splitter.split_documents(documents)
77
- print(f"分割後的文檔數量:{len(split_docs)}")
78
 
79
  embeddings = HuggingFaceEmbeddings(model_name="BAAI/bge-large-zh-v1.5")
80
  print(f"\n成功初始化嵌入模型")
81
 
 
82
  vectorstore = Chroma.from_documents(split_docs, embeddings, persist_directory="./Knowledge-base")
83
  print(f"成功建立 Chroma 向量資料庫")
84
 
85
  retriever = vectorstore.as_retriever(search_kwargs={"k": 2})
86
 
87
  template = """Let's work this out in a step by step way to be sure we have the right answer. Must reply to me in Taiwanese Traditional Chinese.
88
- 在回答之前,請仔細分析檢索到的上下文,確保你的回答準確完整反映了上下文中的訊息,而不是依賴先前的知識,在回應答案中不要提到是根據上下文回答。
89
  如果檢索到的多個上下文之間存在聯繫,請整合這些訊息以提供全面的回答,但要避免過度推斷。
90
  如果檢索到的上下文不包含足夠回答問題的訊息,請誠實的說明,不要試圖編造答案。
91
 
 
10
  from langchain.text_splitter import RecursiveCharacterTextSplitter
11
  from langchain_huggingface import HuggingFaceEmbeddings
12
  from langchain_community.vectorstores import Chroma
13
+ from langchain_community.embeddings import HuggingFaceEmbeddings
14
  from langchain.chains import RetrievalQA
15
  from langchain_community.document_loaders import WebBaseLoader, TextLoader
16
  from langchain.prompts import PromptTemplate
 
74
  )
75
 
76
  split_docs = text_splitter.split_documents(documents)
77
+ print(f"分割後的文件數量:{len(split_docs)}")
78
 
79
  embeddings = HuggingFaceEmbeddings(model_name="BAAI/bge-large-zh-v1.5")
80
  print(f"\n成功初始化嵌入模型")
81
 
82
+ print(f"開始建立向量資料庫")
83
  vectorstore = Chroma.from_documents(split_docs, embeddings, persist_directory="./Knowledge-base")
84
  print(f"成功建立 Chroma 向量資料庫")
85
 
86
  retriever = vectorstore.as_retriever(search_kwargs={"k": 2})
87
 
88
  template = """Let's work this out in a step by step way to be sure we have the right answer. Must reply to me in Taiwanese Traditional Chinese.
89
+ 在回答之前,請仔細分析檢索到的上下文,確保你的回答準確完整反映了上下文中的訊息,而不是依賴先前的知識,在回應的答案中不要提到是根據上下文回答。
90
  如果檢索到的多個上下文之間存在聯繫,請整合這些訊息以提供全面的回答,但要避免過度推斷。
91
  如果檢索到的上下文不包含足夠回答問題的訊息,請誠實的說明,不要試圖編造答案。
92